[Gretl-users] Re: Data File Collection Added : "An Introduction to Statistical Learning"
On Sun, Jan 28, 2024 at 1:49 PM Sven Schreiber
<sven.schreiber(a)fu-berlin.de> wrote:
Am 28.01.2024 um 19:33 schrieb Cottrell, Allin:
> On Sun, Jan 28, 2024 at 12:07 PM Artur T. <atecon(a)posteo.de> wrote:
>> 5. Re-start Gretl to have the dataset available for use.
> Just FYI, if you're running current gretl git you also have the option
> of installing data-file collections via the command-line in the
> console. For example:
>
> pkg install isl.tar.gz
>
> This should also go for the current snapshots.
Thanks, two questions:
- Does the command-line version also require a restart like the GUI
action? (Why actually is a re-start required, wouldn't it be enough to
close the Data Files window?)
No, it's not enough with the current design. When we first open that
tabbed window we search for file collections and cache the
information. Just reopening the window, you get the cached data. That
could be revised (and maybe should be revised) but it wouldn't be
trivial.
- Is the isl.tar.gz thing case-insensitive? In the window I'm
seeing "ISL" (which is good, better than "isl", I believe).
"ISL" is what's shown to the user, based on the "descriptions"
file
inside the package. The filename is all lower-case following the
pattern of most of the tar.gz data-file archives. Whether that makes
any difference depends on the platform. Note that there's something
kinda similar with the Lütkepohl data: "Lütkepohl" shown to the user
but "luetkepohl.tar.gz" for the filename.
Allin